Hen egg white lysozyme (HEL)-specific T cell lines and clones were generated from B6 and BDF1 mice. A variety of clonotypes were found among clones generated at an early stage (1 month) whereas fewer clonotypes were detected after several weeks of culture. Furthermore, a bulk line switched from its initial fine peptide specificity pattern (positive for fragment L2—aa.13-105—and negative for fragment NC—aa.1-17:Cys 6-Cys 127:120-129) to the opposite pattern (negative for L2 and positive for NC), indicating that in bulk lines, besides selection toward oligo- or monospecificity, clones previously silent can emerge after a period of time. Irrespective of early or late cloning, T cell clones could be isolated from three independent T cell lines from different mouse strains that were stimulated by either native or denatured HEL, but not both. Furthermore, 1 clone of 20 from a B6 line, 3 clones of 25 from a BDF1 line, and 1 T hybridoma clone of 10 of B10.A origin lost their capacity to respond to native HEL, yet continued to respond to reduced, carboxymethylated HEL or cyanogen bromide-cleaved, unreduced HEL. These results suggest that T cells may produce activation signals for efficient processing of native antigen.

The period of intellectual history of immunology that we wish to treat began some 30 years ago, in 1958, when the features of immune tolerance had already been described (Billingham, Brent & Medawar, 1953), but mechanisms remained difficult to fathom. The subsequent 30 years in quest of the key(s) to tolerance can be roughly divided into the pre-T-cell suppression half, and the years from 1973 to 1988 in which the major rival to ideas of clonal deletion or abortion has been the T-suppressor cell.

To investigate the mechanism underlying neonatal T cell tolerance, we used synthetic peptides to induce tolerance. We found that induction of tolerance to one determinant on a 23-amino acid peptide did not affect the response to an adjacent determinant on the same peptide. There was no evidence of suppression of the response to the second determinant. Furthermore, even small peptides near the minimal size for a determinant, which would be very unlikely to possess a suppressor T cell-inducing determinant as well as a proliferative T cell-inducing determinant, could induce tolerance. These studies provide in vivo experiments supporting clonal inactivation as the mechanism of neonatal tolerance to immunogenic peptides.

The mechanisms underlying T-lymphocyte tolerance induced in neonatal mice are still unknown. It is unclear whether the tolerant state is the result of inactivation of T cells on exposure to antigen during development or of active suppression by other T cells specific for the same antigen. To distinguish between these two hypotheses, we have analysed the specificity of tolerance to three cytochrome peptides which differ by only a single amino-acid substitution in the epitope recognized by proliferative T cells. The peptides stimulate proliferative responses which are highly specific with minimal cross-reactivity. As antigen-induced clonal inactivation would address the same cells normally activated by that antigen, the specificity of tolerance should exactly match that of the proliferative response to the antigen, and each cytochrome peptide should induce tolerance to itself alone. Conversely, as T-suppressor (Ts) and T-proliferative (Tp) cells almost invariably seem to recognize distinct, non-overlapping determinants on protein antigens, suppressor-mediated tolerance should not be affected by substitutions in the proliferative T-cell epitope. Tolerance would depend solely on the existence of a shared suppressor determinant, so each cytochrome peptide should induce cross-tolerance to the others. We found that the specificity of tolerance matched that of the proliferative response: each peptide induced tolerance for itself but the response to the variants was unaltered. This result strongly supports the hypothesis of clonal inactivation as an important mechanism in induction of neonatal tolerance.

Whether T cell tolerance represents direct inactivation of antigen-specific T cells via recognition of antigen plus major histocompatibility complex, or via T suppressor (Ts) cells, or a combination of these mechanisms, remains to be clarified. This problem was investigated using a novel approach based on the finding in several systems that T helper/proliferative (Th/Tp) cell-inducing antigenic determinants are dissociable from Ts cell-inducing determinants. Thus, peptide probes containing known sites that stimulate T proliferative activity, as well as peptides from distinct sites assumed to bear Ts-inducing determinants, were used in studying hen (chicken) eggwhite lysozyme (HEL)-tolerant mice. The clear prediction from clonal deletion model is that Th/Tp response potential to short peptides in the tolerant mouse would not exist, while regulatory suppression models predict the coexistence of antigen-reactive cells and antigen-specific regulatory cells that prevent their expression. Adult mice, treated with 2 mg HEL in saline, were tolerant to HEL in complete Freund's adjuvant (CFA). Latent T cell proliferative responses could be revealed to determinants within two HEL peptide probes, which lacked the amino-terminal region of the molecule. This responsiveness suggested two conclusions: first, Ts cells directed against the amino terminus of lysozyme exist in the tolerant genetic responder B10.A; second, these Ts regulate the activity of functional antigen-reactive T cells directed against epitopes elsewhere on the molecule, but only in the presence of the complete molecule, HEL. Examination of neonatally induced tolerance did not reveal any latent responsiveness, supporting the hypothesis that clonal deletion or anergy is the relevant mechanism in this situation. Possible reservations in these explanations of the two tolerant states, plus analysis of the more complex "split tolerance" resulting from 20 mg HEL in saline treatment in adults, are discussed. The approach of dissociation of proliferation-inducing determinants from suppression-inducing determinants clarifies our understanding of the tolerant state and holds promise for more definitive exploration of mechanisms of T cell tolerance.

Using synthetic peptides as antigens, it was found that T cell clones of a given haplotype specific for 13-16 amino acid peptides could be clearly distinguished by the varied influence of amino acid substitutions on recognition. This was true for different antigenic determinants within peptides 81-96 and 74-86 of hen egg-white lysozyme, recognized in the context of the I-Ab and I-Ak molecules, respectively. Considerable complexity was demonstrated in the induced T cell repertoire specific for apparently single determinants, which implies that diversity of T cell recognition approaches that for B cells. The implications of the degeneracy of T cell recognition are discussed in the context of mechanisms through which Ia molecules restrict recognition and theories of Ir gene defects.